Hoboken, N.J., March 12, 2026 — When natural disasters or extreme weather events hit, delivering aid quickly and efficiently to those affected is crucial. Humanitarian relief efforts commonly rely on the combination of trucks and drones as a "tag team" to solve the biggest challenge in a disaster: damaged or otherwise inaccessible roads. The trucks act like mobile headquarters, hauling massive amounts of food and water as close to the site as possible. When a bridge is washed out, or roads become flooded or impassable, the drones take over, doing the so-called last-mile delivery, flying over obstructions to deliver urgent supplies like water or medicine where needed.
"This ensures that aid reaches even the most isolated and inaccessible locations," explains Stevens Associate Professor Jose Ramirez-Marquez , who studies disaster recovery and resilience. "The drone can come back, and then it can be resupplied, and then deliver aid over and over again. This collaborative approach combines the strengths of both vehicles, where trucks handle the bulk transportation of goods while drones extend the reach to remote or difficult-to-access locations."
The truck-drone team can significantly speed up aid delivery, but Ramirez-Marquez sought to improve the process further. He wanted to build a mathematical model that would minimize the time it takes for the last person to receive assistance. In other words, instead of just minimizing the average delivery time, the model would aim to shorten the time difference between the earliest and latest deliveries as much as possible. "That way, aid is spread more evenly over time," Ramirez-Marquez explains. "And it's also a fairer way to deliver aid."
Together with Stevens Teaching Assistant Professor Nafiseh Ghorbani-Renani and PhD candidate Ramin Talebi Khameneh Ramin, Ramirez-Marquez used AI and machine learning tools to develop an aid-distribution approach that focuses on optimizing service fairness, workload balance, and minimizing costs such as total distance traveled or operational expenses.
"We used the so-called evolutionary algorithm, because it evolves from one generation to the next," Ramirez-Marquez says. "With each iteration it tells us, 'oh, I found this other solution, and I found this better solution.' At the end, we look at all the good solutions and say, 'you know, this is the best solution we found.'"
To test their system, the team simulated two disaster aid delivery setups, in urban and rural settings. For their first location, the team chose Hoboken, NJ, where Stevens is located, a region with a history of severe flooding, particularly during Hurricane Sandy in 2012, when storm surge inundated the city. For the rural location, the team used a more recent example of the 2025 floods in Hopkins County, KY, which is crisscrossed by multiple creeks and low-lying roadways, making it highly susceptible to flash flooding.
In the second scenario, the team also added an additional component of potential disinformation, which could lead to misleading info about aid requests. "This framework does not infer or detect disinformation," clarifies Ramirez-Marquez. "Instead, it evaluates how prioritization strategies can safeguard equitable access when information may not always be correct in the disaster aftermath."
Using the data generated from flood maps and the location of the aid dispensing points, the team created several scenarios, each of which included a varied number of locations where aid needed to be delivered. Then, the team used the algorithm to generate the most optimal routes for the truck and drone aid delivery for all locations. The team outlined their method in the paper titled Multi-objective optimization of a truck–drone delivery system for fair and efficient humanitarian logistics under disruption and disinformation , published in the journal of Computers & Industrial Engineering in March 2026.
The idea is that if floods or other natural disasters happen again, the emergency responder teams would be able to use the algorithm to generate the most optimal routes minimizing the time it takes for the last person to receive aid. They may also need to adjust it as the situation develops and more requests for aid come in. "They may need to re-run this algorithm as these events unfold and people's needs change," Ramirez-Marquez explains.
The next step would be to work with a municipality to do a hypothetical test run, says Ramirez-Marquez. "The algorithm is ready to be used, now we just need to test in real world settings."
About Stevens Institute of Technology
Stevens is a premier, private research university situated in Hoboken, New Jersey. Since our founding in 1870, technological innovation has been the hallmark of Stevens' education and research. Within the university's three schools and one college, more than 8,000 undergraduate and graduate students collaborate closely with faculty in an interdisciplinary, student-centric, entrepreneurial environment. Academic and research programs spanning business, computing, engineering, the arts and other disciplines actively advance the frontiers of science and leverage technology to confront our most pressing global challenges. The university continues to be consistently ranked among the nation's leaders in career services, post-graduation salaries of alumni and return on tuition investment.
